I Don't Read Charts. I Read People.
A system for tracking 30+ crypto KOLs in 15 minutes a day — so you never miss when a smart mind changes direction.
A system for tracking 30+ crypto KOLs in 15 minutes a day — so you never miss when a smart mind changes direction.
Most people invest by reading charts or on-chain data.
I invest by reading people.
Charts tell you what happened. People tell you what is about to happen — if you know how to listen.
Not to what they say today.
To what they changed.
A fresh opinion is interesting.
A changed opinion is useful.
A pattern of changed opinions across different circles — that is signal.
Investing is Reading People
I have been in crypto since 2017. Through two full cycles, every major move I caught started the same way: not from a chart breaking out, but from someone I trusted changing their tone.
X-Hunt gives you a profile card. Arkham shows you their wallet. SCREENER shows real-time trades.
None of them answer the question I care about:
How is this person’s thinking changing over time?
That is the question that matters. Because markets do not move when everyone stays consistent. They move when smart people quietly update their map.
What I Track
For each KOL, I care about three things:
Stance — bullish, bearish, cautious, aggressive. Not what they say, but where they stand.
Framework — the logic behind their position.
- Luke Gromen sees everything through the oil→Treasury→liquidity chain.
- Hayes tracks the MOVE index as a money-printing trigger.
- qinbafrank connects TMT, macro, and crypto through a unified lens.
Knowing someone’s framework lets you predict how they will react to new events.
Temperature — emotional tone over time. Getting more cautious. More aggressive. More patient. More urgent.
I keep a file per person. Not a spreadsheet of hot takes — a living document of how someone thinks and when that thinking shifts.
How This Helped Me
Case 1: The Hayes Flip (March 5–8)
March 5, Arthur Hayes posted:
“BTC hasn’t decoupled yet from US SaaS tech companies. It could be a dead cat bounce. Be patient.”
My system tagged this: ➡️ continuation. Still cautious.
March 8, same person:
“Kabloom… Hold on to your butts!”
Tagged: 🔄 stance shift. Three days, cautious to aggressive.
Same day, Alex Becker — a completely different circle — posted: “The most hated rally has begun.”
Tagged: 🤝 consensus resonance.
If I had been scrolling Twitter casually, I might have seen one of these posts and shrugged. But because I had both people’s previous positions on file, I could see: this was not just two people being bullish. This was two people who were recently cautious, shifting direction in the same week.
What I did: upgraded my attention level on BTC. Not a trade signal — a research trigger. I spent the weekend digging into what changed.
Case 2: The Oil→Treasury Chain (March 3–9)
qinbafrank had been tracking the Iran situation with a clear framework: the conflict’s duration, not its existence, is what determines oil/gold/equity direction. On March 3, he flagged Hormuz Strait closure risk as the market’s “true eye.”
Luke Gromen, from a completely different angle, posted on March 9:
“Iran doesn’t have to beat the US military; it just has to beat the UST market.”
Two persons, different hemispheres, different languages, converging on the same insight: this is not a military story, it is a Treasury market story.
What I did: I had already read qinbafrank’s analysis days earlier. When Gromen independently arrived at the same conclusion, that convergence told me this was not one person’s speculation — it was a structural read that multiple credible minds reached independently. This shaped how I thought about my portfolio’s macro exposure.
Case 3: The Signal in Silence (Raoul Pal, all of March)
While Hayes flipped and Gromen escalated, Raoul Pal kept posting about “Economic Singularity by 2030” and “There’s no stopping the tech revolution.” No pivot. No flinch. No mention of Iran or oil disruptions.
I mark that ➡️ continuation.
This might seem like non-information. But when a thoughtful macro voice refuses to engage with the panic narrative, that tells you something about their conviction level. Pal is not ignoring the news — he is telling you it does not change his structural thesis.
Not every signal is a flip. Sometimes the signal is who refused to change their mind.
Four Signals I Trained the System to Flag
- 🔄 Stance shift — same person, different direction vs last week
- 🤝 Consensus resonance — 3+ people from different circles pointing the same way
- ⚡ Silence break — someone who rarely mentions a sector suddenly starts
- ❌ Words vs actions — public tone says one thing, on-chain behavior says another
Why This Beats Scrolling Twitter
I follow 30+ KOLs. Reading all their tweets every day would take hours and I would still miss the pattern.
The pattern is not in any single tweet. It is in the delta between what someone said last Tuesday and what they said this Saturday.
No human brain holds 30 people’s evolving positions in working memory. But a system can. And when it flags “three people from different circles changed direction this week,” that is something I could never see by scrolling.
I do not need thirty perfect calls. I need to notice when a few credible ones change direction around the same time.
What I Built: OpenClaw + 6551 API
I got tired of doing this manually so I built a system that does it for me.
- OpenClaw — an AI agent framework. I run multiple agents to scan KOLs, analyze signals, and draft content from the findings.
- 6551 API — provides structured Twitter data with AI scoring. One API call returns a KOL’s recent tweets, filtered and scored.
Every morning at 8:30 AM, my agent runs a scan across a tiered KOL list:
- Tier 1 (daily): RaoulGMI, CryptoHayes, Arthur_0x, qinbafrank — core macro×crypto voices
- Tier 2 (rotating): AI/Crypto builders, macro analysts, contrarian thinkers
- Tier 3 (weekly): Broader check-ins for big moves only
At 8:45 AM, a second agent reads the raw data, extracts substantive views, compares against historical stance files, and flags signals.
No manual scrolling. No missed updates. Under 15 minutes.
How to Build Your Own Version
Step 1: Pick Your List
Start with 5. Not 30.
- 1 macro voice (shapes your risk posture)
- 2 voices from sectors you hold
- 1 voice you usually disagree with
- 1 voice outside crypto entirely
Step 2: Set Up Your Tracking
Option A: Manual (works fine)
| Date | KOL | Key Take | Tag | vs Last Week |
|---|---|---|---|---|
| 3/5 | Hayes | ”BTC hasn’t decoupled, be patient” | ➡️ | — |
| 3/8 | Hayes | ”Kabloom, hold on to your butts” | 🔄 | cautious → bullish |
| 3/8 | Becker | ”Most hated rally has begun” | 🤝 | resonates w/ Hayes |
15 minutes a week. Fill it in on Sundays.
Option B: Automated with OpenClaw
- Install OpenClaw
- Install the kol-stance-tracker skill
- Get a 6551 API token
- Configure your KOL list and schedule a daily scan
- Your agent handles the rest
Step 3: Ask One Question Every Weekend
Who changed, and why now?
If nobody changed, that is data too. Stability in a volatile market is its own signal.
The Compound Effect
After a few weeks of doing this, something shifts in how you consume information.
Your timeline stops looking like chaos. It starts looking like a series of early tells.
You stop asking “What are people saying today?” You start asking “Who changed, and why now?”
Over time, you also learn something more valuable: who updates early, who updates late, who overreacts, who stays steady. That is not just useful for markets. It is a trust map — built from data, not impressions.
Stay Invested in the Game. 🐍